TiF's just higher quality. And you want to be doing the OCR on the highest resolutiion scan you can get.
What I do is take all those TIF files then create a PDF with Adobe Acrobat (not just the reader software, the real program). After that I might OCR it with Adobe's internal one, or use the one that comes with the OpticBook....
But I keep the TIF files around until I'm completely finished with the process of converting - ideally, where you want to be is something that is pretty open like a Word or HTML type file.
When it's all done, then I delete them. BUt there's no point in converting them to jpeg's at all.