(At @jakebeal's request, these are my comments turned into an answer.)
Is there a particular reason why you can't do John (2010, pp.10-12)? I would hope most readers could understand that any information you were citing came from one of those 3 pages and could go and search for it themselves.
In my mind it's a matter of readability over specificity. Part of that comes from my background in computer engineering - I subscribe to the belief that when writing code, it's more important that a human can read the code than its conciseness (with exceptions, of course). Similarly, when writing a paper, I personally think it's more readable to combine the 3 pages into a single citation and only interrupt the reader once.
Imagine critiquing an author's argument that spans 10 pages - it becomes a mess after a while:
Smith (2nd Age, pp. 389) claims in his book "The Art of Dragon-Ryding" that red dragons are the easiest to ride. He backs this up by pointing out their smaller stature (pp. 390), their powerful hindlegs (pp. 391), their intelligence (pp. 392-393), their natural resistance to werewolves and other such creatures (pp. 394), and so on.
Much simpler would be to use a single citation and allow readers to go searching for specifics - so long as they know it's not YOUR idea and where they can go to find the idea. Note that (at least in my opinion) this is easier to read with little if any loss of accuracy:
Smith (2nd Age, pp. 389-94) claims in his book "The Art of Dragon-Ryding" that red dragons are the easiest to ride. He backs this up by pointing out their smaller stature, their powerful hindlegs, their intelligence, their natural resistance to werewolves and other such creatures, and so on.
As a final note, remember that not every field of academia uses the same citation style (and not even every subfield within a field). When citing papers or books, the standards may change even from journal to journal within a particular subfield. So when in doubt, talk to the authority in charge - be that a course instructor, a journal editor, a conference organizer, etc.