I kind of doubt it's Leptophlebiidae, because they don't have such mottled wings. At least not the genera I'm familiar with. It's possible, but unlikely, that a South African genus in that family might have mottled wings while the North American genera do not.
I doubt it's Oligoneuridae, because they're supposed to have this characteristic venation:
Forewing venation greatly reduced, apparently only three or four longitudinal veins behind R1.
This certainly doesn't fit that characteristic, which I found photographed here
. However, that characteristic is from a North American key, so it might be unique to the small number of North American species, and perhaps African Oligoneuridae don't share that trait. It seems unlikely, but it's possible.
According to Roger's website, the adults of one North American Oligoneurid, Lachlania
, have just two tails. Usually that's a consistent number throughout a family, but I'm not positive it's always the case. But it seems like a strike against that guess.
I found a page listing the mayflies of South Africa
It seems there aren't a lot of options other than what you guessed. Heptageniidae might make sense, except that it has 3 tails, and they don't. There are two families in the superfamily Ephemerelloidae that I've never heard of, so I wouldn't rule them out, but it just doesn't quite have the "look" of that group. I also wouldn't rule out Ephemeridae, but again the "look" isn't quite right. Other than those, the two families you suggested are the only viable options.
I'm really stumped. If I had
to guess right now, I'd guess that it's a Leptophlebiid with pretty different wings from the species I'm familiar with.