As a student—albeit an undergraduate student—who's been to talks that were well over his head and as a student who has presented math that was intended for a general audience but ended up going over people's heads, I'm really interested in this question. For what it's worth:
I'm going to the JMM in Boston this year and I'm entertaining the idea of emailing people who's talks I'd like to go to and ask them to email me their paper so that I can get a preview of coming attractions. I'm thinking that reading the paper will be as helpful to a deeper understanding of presentations as reading the chapter does before it's covered in class. If I were in your position, I would try to get my attendees access to some literature ahead of time. I don't know how realistic this is, but the idea is that this is one way for people to find out what tools are going to be used in the talk. If they need to make a trip to the library (or Wikipedia) before understanding the material, they can do that before the talk.
I also like the idea of recorded lectures. That way if I don't understand something, I can go back, rewatch it, pause, work out the computation, whatever and start it going again.
And I think the link in the first answer has an extra character in it and it should be: http://www.ams.org/programs/research-communities/mrc-12
