Jim Hu gives us another reason for scientists to consider publishing in open access journals:
Sometimes I'd like to view your papers while I'm off campus and at a study section. Of course, if you're one of my grants, I have already accessed your paper from home. But if I can't access the paper from the NSF building in Arlington, or from a hotel where an NIH panel is meeting, I can't use information from it to argue against some other panelists misinterpretation of what you did.
Publishers who are reluctant to go full open access might consider providing a means for NSF and NIH reviewers to get free access.
To be fair, universities generally offer some kind of off-campus access to electronic journals for their faculty and students. However, these systems often require some sort of premeditation (i.e. signing up to use them while you're still on campus), can be prone to technological hiccups, and/or can be cumbersome to use. In a fast-paced environment, being able to access the full text of a publication with just a few mouse clicks can be very advantageous, so Hu makes a point worth considering.