Here's a very simple view about reasons. It makes just two claims:
(O) You have (objective) reason to do that which would best fulfill your desires in actual fact.
(S) You have subjective reason to do that which is most likely to fulfill your desires, given your beliefs.
This makes it clear why true beliefs are so worthwhile. They help us to close the gap between subjective and objective reasons, allowing us to notice (and act on) the best reasons. But this benefit would accrue to any theory which made a similar distinction between objective reasons and belief-relative 'subjective' ones. So the real issue here is with the substance of (O) - that we only have reason to fulfill our desires. (See also my related post on the desire-fulfillment theory of wellbeing.)
I guess the core assumption behind this view is that a reason is necessarily something that would guide the action of an informed agent. But according to the belief-desire theory of action, all action aims at desire-fulfillment. So the agent would treat as action-guiding only information that was relevant to the fulfillment of his desires. If you presented him with a 'reason' which appealed to some end that he did not share (i.e. desire), then he would not consider it a reason to act at all.
Any objections? Could there be reasons that informed agents would not care about?