The article cited above indeed looks like a good read. As has been said, *all* resistors produce noise which is proportional to their resistance & temperature - the Johnson/Nyquist noise referred to in #3. They will also produce "excess" noise of a nature that depends on their construction - carbon composition R's are notoriously bad in this respect.
I guess the question is really, are your cheap resistors going to be what they say they are? For instance, could they be sold as metal film when they're actually carbon film? To be honest I don't know how you'd tell. If you break them open certainly it's be easy to identify certain types, but telling the others apart just by looking at them might be tricky - I've never tried.
Same story for the transistors really, although it should be easier to measure the noise of them. As for what makes them "low noise", as audioguru says... a bit of luck, and some magic I don't pretend to understand...