I think the answer to this question is Yes and No (as is generally the case for complicated questions, I suppose).
What we see in animals is relevant to questions about human speciation … unless you consider genetic engineering, in which case the limitations of natural selection or drift are not important. So in the real world, I don’t think natural processes have much to do with what might happen with humans in the future. Still, here are some interesting notes, most of which I didn’t get a chance to share at the speciation panel at WorldCon:
1. Sometimes speciation can happen surprisingly fast. The cichlids of Lake Malawi and Lake Victoria and so on speciated VERY VERY FAST. We know this because the lakes themselves are not very old.
In something like 10,000 years, these lakes filled up with thousands of distinct species. Why did cichlids radiate so fast in these lakes? The answer appears to be twofold:
a) Extremely intense feeding competition drove niche partitioning and rapid changes in jaw morphology. This all got a boost because of the structure of the lake edges, which are broken up into lots of fairly separate bays, so there were a lot of chances for more-or-less allopatric (separate) populations.
b) Lineages which are sexually dichromatic speciated like crazy. Species where the males and females basically look alike didn’t. This shows the importance of sexual selection in driving rapid bursts of speciation. In cichlids, apparently female preferences shifted and diverged rapidly, driving very quick speciation in parapatric (abutting) populations.
2. However, you can and do get speciation in highly mobile, sympatric (occupying the same area) populations. Both African elephants and orcas have speciated (or are in the process of doing so) even though their populations ought to be able to mingle very easily, and even though to an outside observer the populations don’t look all that morphologically distinct.
Orcas seem to be undergoing speciation based on reproductive isolation of populations that follow learned feeding preferences. (Do you eat seals? Us fish-eating orcas will never associate with you.)
Savannah and forest elephants are not actually as similar as it seems to a quick human glance. Savannah elephants are about twice as big as forest elephants, and they look distinct enough that you can learn to tell them apart at a glance. But it was still a surprise to find out they are as genetically distinct as Asian elephants are from mammoths. Behaviorally, the two species of African elephants are pretty different. It’s the savannah elephants that live in larger families and congregate into what we might call clans.
3. Sometimes closely related species interbreed surprisingly freely, but they are still distinctive species. Savannah baboons (to my mind, all one species, whether they’re called olive baboons, yellow baboons, or chacma baboons) have a continual influx of hamadryas baboon genes where the two species are found together. This is because young male hamadryas baboons are excluded from the one-male-plus-harem type of group that is customary in their species. But they can get accepted into a multi-male-multi-female savannah baboon group, and once there, they sometimes breed, especially once they learn to quit chasing and biting the females. Female hamadryas baboons respond to that kind of treatment by following the males and sticking close to them, but savannah baboon females don’t respond that way at all — they run away from their attacker, often to their males friends for protection.
So when they mature, male hamadryas baboons put their energy into acquiring a harem of hamadryas baboon females, but in the meantime, there is that continual low-level mixing. Yet though the two species have probably been trading genes like that for a good long time, they are still distinctive species. Though who knows; the hybrid zone is a dynamic thing and it’s not clear exactly what’s going to happen in the future.
But I do think this is all interesting — that speciation can be fairly fast, that a low-level influx of foreign genes doesn’t necessarily prevent the formation and maintenance of distinct species, that being highly mobile and even sympatric doesn’t necessarily prevent speciation; that apparently learned behavior can be a potent reproductive isolation mechanism. You can sure see how even aside from genetic engineering, if you scatter humanity across the galaxy, all these factors could be highly relevant.
This would be a really fun topic to play with if designing a far-future SF universe. I must admit, I haven’t decided whether to go this way in my current space opera universe, but I think not. Ignoring the potential for speciation has a long and honorable tradition in space opera, just as much as FTL wormholes. Still, it’s interesting to think about.