Some sites have a custom close reason to handle this kind of cases. For example, on Unix & Linux:
Questions describing a problem that can't be reproduced and seemingly went away on its own (or went away when a typo was fixed) are off-topic as they are unlikely to help future readers.
I doubt that CSTheory gets enough questions of this type to warrant a custom close reason. If a question is about understanding a point in a research paper, then
- either the point is a subtle one that others may miss as well (or the paper omitted a step of the proof), in which case the question is potentially useful to anyone reading the paper and should be answered and remain open;
- or the point is a basic one that the asker happened to miss but where other readers aren't likely to stumble, in which case closing as off-topic with a custom comment plus a comment to explain is fine.
Basically, I agree with what you did, except for one thing: don't use the words “too localized”, which is historical Stack Exchange jargon that most people won't understand. Write something like
This question concerns a basic point and is unlikely to be useful to future visitors, therefore I am voting to close it.