OK, let me preface this by saying that I'm no brain surgeon, I only had a major in psychology... but I think I remember some things from my biopsych class...
Basically, the "reptile brain" things you're talking about mostly happens in the cerebellum. Or, at least, that's the part of your brain that you train when you're training most physical actions that you remember in an "instinctive" way. The classic example is riding a bicycle. Apparently, you could get almost a full frontal lobotomy, so you wouldn't know what a bicycle was, the word for it, or know how to do anything but drool on it if someone put it in front of you, but if you got on it, you could still ride the thing.
Anyways, so my guess is that in Aikido, we're training the cerebellum. It is a process of taking concious control over how our unconcious processes work. And that's pretty darn cool in my book.
Now, as for the rest of it being "Japanese hocus pocus," I both heartily agree and totally disagree. Yes, there is probably no need to have exactly the surroundings we have to learn Aikido. However, at the same time, there are a lot of concepts there that you couldn't train your body to accept without acknowledging them conciously first. From what I have seen, most people don't really have a good concept of how to move while remaining mostly centered. Aikidoka (at least the good ones) do.
Either way, we know that the "Japanese method" of training the brain has worked in the past, and does work for some people. If you can suggest some improvement to that method based on neurobiological research, then by all means, let us know, and I'm sure a lot of people could benefit from it. However, until then, what we've got seems to work at least tolerably well.
Bottom line: "that's great, now get back in the dojo and train!"
The other Ted