Well, first I'd like to say that knowing you plan on doing real world research shows a lot of commitment on your part to do a good job.
But I also think a lot of DM's spend their fair share of time watching shows on history channel, discovery channel, learning channel, PBS specials, and maybe even real life current events with which to grab plots and fantasy campaign ideas. Heck, the travel channel even did a show on the 10 best dungeons!
I think a brief article on how to turn current events into fantasy plots would make a good article for Asgard magazine. However, an accessory book that covers past myths and legends is the sort of thing every fantasy reader already has some grasp of from television, movies, and books. What is it that you hope to add that your average fantasy gamer wouldn't know?
If you're just making a bunch of game stats for NPC's or feats to duplicate mythical abilities I wouldn't be interested at all--too easy to come up with that stuff on my own if I wanted it. Are you going to describe ancient sites?--There is a book in the bargain bin at Barnes & Noble's that I've already browsed through in half an hour, and no longer need. And I have another book that discusses ancient mythical figures as well.
A book on sacred rites of passage or rituals has always interested me. But I actually did my homework and then translated those things into something dramatic for my own campaign. It's not often enough that I have an excuse to have naked dancing women summoning a devil to serve their needs!
I guess all I'm saying is that if you're going to rehash what's already out there I think you would be wasting your time. Your myths book will need to feel new and unique enough to warrant a "I'll buy that!," kind of appeal to your customers, and that means having not seen something like it somewhere else, either on television or in books.
--Painfully
painfully@mindspring.com
PS. [OT, I know but I'm too lazy to make another post] I liked the wild spellcraft book--I use a system that allows for spell failures as well, using % dice and ratings according to how often they've successfully cast the spell in question. Almost nobody will cast a new spell right for the first 10 tries, thus encouraging wizards to spend time practicing spells during their downtime. The flipside is that once they reach a particular proficiency I let them take a metamagic feat for that spell. On the extreme end, should they cast a particular spell successfully for the 10,000th time, I let them pretty much have every applicable metamagic feat for that spell.
I let them fizzle the spells when they fail, but a roll of 01 always gives the worst possible effect I can imagine at the time.
Simple, and not exactly worthy of a book to sell, but I think simplicity is an asset.