You're right, we do understand the pathophysiology of many diseases, and those are the ones that have been mostly eradicated. The major chronic diseases that remain are very poorly understood such as type II diabetes, cancer, cardiovascular disease, and alzheimer's.
I spend a lot of time reading about 'alternative' ideas about these diseases, and many seem promising, but aren't taken seriously by the mainstream. It's definitely possible that they're ignored for a good reason, but I haven't been able to find the reasons yet. This is the biggest problem I've found with trying to be 'critical of everything.' In very few instances do I find myself quickly understanding and agreeing with the mainstream view. Instead, the more I read the more my opinion seems to diverge from the mainstream view. I have made an effort to discuss these issues personally with specialized experts, so they could help point out factors I may be missing, or not understanding correctly. I am a PhD candidate in the life sciences, so I have the opportunity to meet with research professors at my university in person to help clarify my understanding.
Here are two example theories, regarding cancer and cardiovascular disease in particular.
1) The idea that cancer isn't initiated by genetic mutations, but that mutations are a downstream phenomena that results after damage to the mitochondria occurs.
This stems from the initial observation by Warburg, that lack of control over glycolysis is part of the cancer cell phenotype. This phenotype can be triggered by a large number of factors which inhibit mitochondrial respiration including hypoxia. Later it was found that the mitochondria in cancer cells undergo a phenotypic change, where the cristae structure is lost. Nuclear transfer experiments have shown that a 'mutated' cancer nucleus placed into a healthy cell cytoplasm does not exhibit a heritable cancer phenotype. Conversely, a healthy nucleus placed into a cancerous cell cytoplasm does exhibit a heritable cancer phenotype.
Here is a review article covering the evidence for this hypothesis:
Cancer as a metabolic disease: implications for novel therapeutics http://carcin.oxfordjournals.org/content/35/3/515
More evidence for this hypothesis includes the observation that active thyroid hormone levels (T3) are inversely correlated with cancer mortality rates in the general population. T3 is a key regulator of mitochondrial respiration:
Thyroid hormones and mortality risk in euthyroid individuals: The Kangbuk Samsung Health Study. http://www.ncbi.nlm.nih.gov/pubmed/24708095
2) The finding that treatment for hypothyroidism drops cholesterol levels significantly, and virtually abolishes cardiovascular disease without the side effects seen from statins. The late Broda O. Barnes was an experimental endocrinologist and a clinical doctor, and he extensively documented this phenomena in his books and publications.
The idea here is that the central mechanism of cardiovascular disese is a low metabolism which inhibits cholesterol clearance from the blood via reduced steroid hormone synthesis, and reduced bile synthesis. The pathophysiology of cardiovascular disease begins with a long residence time of cholesterol particles in the blood, resulting in their oxidation. This can be reversed by any strategy that restores a normal (higher) metabolic rate: a carefully designed diet and/or thyroid hormone supplementation.
Here is a good introduction to this idea:
The Central Role of Thyroid Hormone in Governing LDL Receptor Activity and the Risk of Heart Disease http://blog.cholesterol-and-health.com/2011/08/central-role-of-thyroid-hormone-in.html
I am not insisting that these ideas are correct, or are some sort of 'well proven answer' to these diseases. I'm just pointing out that they seem promising, but are relatively ignored. If they prove accurate, much of the mainstream research on these phenomena would seem to be barking up the wrong tree.
You might notice that both of these examples are essentially the same theory. This is an appealing concept to me: most age-related chronic diseases may be centered around a common process of age related impaired mitochondrial function and/or improper hormonal regulation of mitochondrial function. Insufficient chemical energy (ATP) to fuel normal biological function would have widespread consequences, and could present as a diverse array of seemingly disconnected symptoms. I'll admit, this sounds somewhat like a modern molecular version of vitalism. However, unlike vitalism it makes specific testable predictions, and involves a very specific mechanism. It's also consistent with the 'free radical' and 'tissue peroxidizability index' theories of aging, which involve (among other things) progressive oxidative damage of unsaturated fats (such as cardiolipin) in the mitochondrial inner membrane.
Suppose you distrusted everything you had ever read about science. How much of modern scientific knowledge could you verify for yourself, using only your own senses and the sort of equipment you could easily obtain? How about if you accept third-party evidence when many thousands of people can easily check the facts?
My purpose with the question isn't to cast radical doubt on science; rather, it's an entertaining game of trying to understand how we know what we know. Thinking through these sorts of questions also helped me notice interesting things in the history of science that I hadn't previously focused on. It might also be of interest from a science education perspective.
Some things are much easier to check than they used to be. As late as the 19th century, there were people who were publicly skeptical about the curvature of the earth. Skeptics and scientists did careful measurements (notably the Bedford Level Experiment) to observe the earth's curvature. Today, you can verify it by phoning a friend a few time zones away and noticing that the sun reaches the zenith at steadily later times as you move west. This only makes sense if the earth is curved.
Some things are still hard to check. I don't know an easy way to show that the Earth orbits the Sun. The direct way to show it would be to measure stellar parallax. But even the closest stars have a parallax of less than an arcsecond. My understanding is that very few amateurs are able to take measurements with that level of precision.
Some things are surprisingly easy. There are lots of easily accessible demonstrations of quantum phenomena. For example, a ten dollar spectroscope will show you that an incandescent light bulb has a continuous spectrum, and that LEDs and fluorescent bulbs don't. Bright-line spectra are very much a quantum mechanical phenomenon -- it's a sign that the atoms in the light source have fixed energy transition levels. Spectroscopy was one of the key early lines of evidence for quantum mechanics, and it blows my mind that it's something you can just see whenever you want, with a negligible equipment cost.
Pretty much all of modern chemistry and solid state physics rests on a quantum foundation, and you can test a great deal of chemistry pretty easily. If you are in doubt that water is a bonded compound of two gasses, you can do the electrolysis very easily yourself. You can observe the periodicity of chemical elements yourself if you buy alkali metals (don't try this one at home!). If you are willing to accept slightly indirect evidence, the entire semiconductor industry is about precisely controlling the conductivity of impure silicon, and this would make no sense if quantum mechanics weren't a reliable guide to electron energy levels in the solid state.
I don't feel quite as qualified to play this game for biology. I imagine that antibiotic resistance is a well-enough documented case of evolution through natural selection to serve at least as a proof of concept. DNA sequence comparisons across species are emphatic evidence of taxonomic trees, if you trust the scientists not to be part of a vast conspiracy.
It feels almost impossible that it's easier to see quantum mechanical effects than it is to see that the earth orbits the sun, but it does seem that way.
Some questions: