And it's not as if his books stand a chance of converting people who are already religious...the dismissive attitude that comes through in his writing is exactly what WON'T make people really change their minds.

You (and a lot of people) say that, but I haven't seen evidence presented that they don't work - just people's models of other people.

However, I note David Colquhon's discussion of how he killed the study of homeopathy at several UK universities:

Dr Baggini, among others, has claimed that the “new atheists” are too strident, and that they only antagonise moderate atheists (see The New Atheist Movement is destructive, though there is something of a recantation two years later in Religion’s truce with science can’t hold). I disagree, for two reasons.

Firstly, people like Richard Dawkins are really not very strident. Dawkin’s book, The God Delusion, is quiet and scholarly. It takes each of the arguments put forward by religious people, and dissects them one by one. It’s true that, having done this, he sets forth his conclusions quite bluntly. That seems to me to be a good thing. If your conclusions are stifled by tortuous euphemisms, nobody takes much notice. Just as in science, simple plain words are best.

The second, and more important, reason that I like Dawkin’s approach is that I suspect it’s the only approach that has much effect. There is a direct analogy with my own efforts to stop universities giving BSc degrees in subjects that are not science. Worse, they are actively anti-science. Take for example, homeopathy, the medicine that contains no medicine. I started by writing polite letters to vice chancellors. Usually they didn’t even have the courtesy to reply. All efforts to tackle the problem through the “proper channels” failed. The only thing that has worked was public derision. A combination of internal moles and Freedom of Information Act requests unearthed what was being taught on these courses. Like Westminster’s assertion that “amethysts emit high Yin energy”. Disclosure of such nonsense and headlines like

Professor Geoffrey Petts of the University of Westminster says they “are not teaching pseudo-science”. The facts show this is not true

are certainly somewhat strident. But they have worked. Forget the proper channels if you want results. Mock what deserves to be mocked.

Or, as Mencken put it decades ago:

One horse-laugh is worth ten thousand syllogisms. It is not only more effective; it is also vastly more intelligent.

I suspect your true rejection is the claim that Dawkins is "unnecessarily critical". Unfortunately, this usually means "critical at all".