@thermz: It's not strange, it's usually an X-Y problem like amon said. Most questions with "very narrowed scope" are arbitrarily narrowed because the author has already decided on a solution, implemented it partially, and gotten stuck on the rest, and just wants a jump start instead of having to backtrack. Sometimes there really are special circumstances (mostly legislative/regulatory requirements like PCI, SOX, that kind of thing) but those are rare relative to X-Y problems.
In fact I'd hazard a guess that nearly half of all questions closed as "too broad" (on this site) are specifically questions where the OP just hasn't even made an effort. "Too broad" is the symptom, not the cause.
By saying that there are existing close reasons that fit as a primary argument, I think you're essentially going against your own rationale for creating those reasons in the first place. Sure, you could use one of the existing reasons, but the point is to help the person by being as specific as possible about what the problem really is, and it would be much kinder to close questions like this one as "needs more research effort" than by "too broad" with a flurry of downvotes.
@Rachel: A good question can have more than 1 good answer, but not an infinite number of equally-good answers, otherwise it's not a question at all. So far, neither answer presents any really convincing rationale for why it's better than any other answer, nor does the question itself ask for such a justification or even give any basis on which to form a justification. It's another "what language" poll, plain and simple.
@Renesis: Sometimes questions fall under multiple close umbrellas. I've seen questions that all five can apply to (most of them being career questions). I think you're fighting an uphill battle claiming that the question was constructive either - what problem is it solving, exactly? But the specific close reason is honestly not as important as the fact that it's closed. If it's closed by community vote, then the official close reason might not even accurately reflect how everybody voted. Much ado about nothing, IMO.