I'm not actually sure that the diet, as you've written it, would work even if the theory were complete. You say that an association is formed between flavor X and calories and that association with X controls the set point. But why would X be the dominant factor, when you're already eating flavors A, B, through to W whose contribution to the set point got you to the weight you started at? Does the book elaborate on how the association works?
The first example to come to mind is Richard Halley from Atlas Shrugged, but I don't remember the book all that well.
How would you know that you didn't both erase the memory of some event and erase the memory of erasing the memory of that event? The more you commit yourself to not tampering with your memories, the stronger the cognitive dissonance will be at having tampered with your memory.
I think what he means is that addiction is a physiological condition. A number of drugs leave withdrawal symptoms if you try to stop taking them (alcohol, opiates, etc), but this sense of the word addiction is medical only. There are lots of other things (like computer and gambling addictions) that have only a psychological dependence [so are learned] and could be helped with memory blockers.
As an aside, it's a very low-resolution mindset that disregards an entire article because of one perceived mistake.
Have they mastered the art of getting ECA in countries where Ephedrine is a controlled substance?