Generating Classical Groups over Finite Local Rings asks a question that, according to the poster's own 'answer' https://mathoverflow.net/a/19098/2383, is not what was actually meant. I edited the question to reflect the stated intention (changing the words "semisimple elements" to "tori").
One rejection said:
This edit does not make the post even a little bit easier to read, easier to find, more accurate or more accessible. Changes are either completely superfluous or actively harm readability.
I have the impression that this rejection was based on the idea that I was trying to substitute a synonym for "semisimple elements" merely to improve the wording, whereas actually I was (intentionally) changing the meaning (so that it agreed with the poster's stated intentions—5 years' interval suggesting that he or she was not going to make the edit him- or herself).
That brings me to the next rejection, which said:
There is already an answer; rather than edit out the question that was answered, you should add the clarification so answer is not dinged as wrong.
This seems like a very clear action plan to me, and I agree that doing this is better than making my original edit. However, given the first rejection, I am reluctant to try again to make this edit in case there is any penalty for appearing 'argumentative'. Is it reasonable for me to try again?
EDIT: I was too slow, and, in the meantime, the post was edited in a better way, simply inlining asm's clarification. I didn't pay enough attention and submitted my in-progress edit anyway, but I guess that it will (properly) get re-rejected.