An Alk over 16 dKh seems pretty damned high to me. And I corrected it to be LOWER with water changes. pH was also incidentally raised to 8.25 with said water changes.
7.9 is for many larval species a perfectly acceptable pH. It may not be so for these gobies, but I have no reason to think they are different in this respect. Its not as if the pH was 7.5 or some such ridiculously low number. Since I am using Clor-Am-X, I do not need to worry about a pH of 8.25, which is pretty high.
OTOH, the dying fish were skinny and wasting away; I covered the back of the tank and the survivors immediately moved to the center of the tank and started eating. I could literally see cause and effect within minutes. They don't get fat like the clowns but at least they have bulges in their middles now.
THE cause for improvement? I can't say. But it is one which I can specifically identify, whereas everything else is speculation.
When the O2 test arrives (which is backordered, of course) I can check that item, but I really have no concerns about a pH of 7.9, in fact, its pretty close to what I'd like to see provided the O2 levels are okay. My error -- if indeed it was the fatal item -- was in overusing the buffer instead of addressing the root causes of the pH drop, not the pH itself.
IMO.