It's not just a question of what you are recording, but also what microphone you are using. Preamp noise specs are usually listed as equivalent input noise. That means if the noise rating is -90dBV (it must also have a reference, the "V" in this case means 1V, another possibility is "u" which is 2dB higher at the same voltage), that would be like using a microphone with -30dBV/Pa sensitivity, and a 34dBV noise rating.
That would be a noisy microphone, on par with very small cardioid lapel-type mics, so you might feel that -90dBV is an undesirable noise rating for a preamp, and truthfully in a modern preamp it is. To avoid that problem, you'd need to have a hot enough mic and source to get your signal far above the noise. That probably means no ribbon mics for anything except guitar amps or drums, for example.
But first be sure that the noise ratings are really and truly comparable. There are a few other ways to state a noise measurement, although it befuddles me why any manufacturer would punish their specs by quoting them in such a manner. Another concern is weighting; noise specs are often listed using A-weighting (which personally I feel is appropriate), but other manufacturers might cite unweighted noise readings, which is like a 6 or 8dB penalty.
Another question is what gain the manufacturer measured noise, and how that affects the preamp's noise rating. You might have noticed that some preamps get suddenly noisier at the top of their gain range. Others don't, and will show similar noise at any gain setting. If you are concerned about preamp noise, that probably means you are concerned about high gain settings, so that is the relevant statistic to know.
