Regarding the creation of fertile offspring being a criteria for a species, I believe that this is only for higher organisms. Mammals for sure, but I'm not sure how far the line that rule applies. I believe that cage-kept birds will hybridize and create fertile offspring. Not sure about reptiles or amphibians.
Huge variation within 'classes' of animals, but species within classes, orders and sub-orders generally cannot interbreed unless they come from within the same 'family' or 'genus', and even then its generally uncommon/impossible (see insects/beetles).
Point I was making is that many of the lake cichlids just aren't very genetically distinct. Anybody can and will basically breed with anybody.
Know what Brussels sprouts, kale, kohlrabi, cauliflower and broccoli have in common? All the same species. Thats what a little hybridization can get you.
Suppose the good news is that if the Rift Lakes ever flatline that it's possible for it to take only a mere 10,000-15,000 years to rediversify. Although if you start with Nile Perch or Tilapia as the progenitor species it could turn out to be a very different sort of deal. Probably safe to say that some of the Rift cichlid species weren't around 500 years ago because the genes being expressed now hadn't been isolated and compounded through selection to differentiate/distinguish them as a 'new' species.
I'm sure I have no idea but the whole dance is fascinating in the extreme.