mp3s at 128 kbps are adequate–not even good–sound quality. Leave them alone because any conversion will either make them much bigger (lossless like FLAC) or worse (lossy like ogg).
You need to start with a good-quality source before you compress it. I suggest you do a test. Get a CD you know fairly well and convert one song a bunch of ways. You can use the free Media Coder to make mp3s at various bitrates (128, 192, 256, 320), ogg at various bitrates, and FLAC.
Put them all onto the Sansa and see how much difference it makes through your headphones into your ears. Try to randomize them so you don’t know which one you are listening to–so your ears are making the decision, not your brain reading numbers or format types. Then pick the one that you think strikes the best balance between sound quality and file size.
Audiophiles drive themselves insane with this stuff because they look at numerical readouts and see that ohmigod there’s a tiny bit of lost information between 14,500 and 14,750 Hz in the left channel. Can you hear that? Can I hear that? Probably not.
You may well notice an improvement in sound quality between mp3 at 128 kbps and mp3 at 192 kbps. But between mp3 at 256 kbps and a FLAC file that’s 5 or 6 times larger? Highy unlikely.