^
Some low level hiss during quiet moments is far better than clipping distortion during the loudest parts. Just right is Goldilocks, but too low is safer than too hot.
For an audience recording - especially a rock and roll recording - the noise floor of the environment - crowd moving, chatter, HVAC etc - will far surpass that self noise number.
^this.
The only way it may backfire on you is if you record at crazy low levels and have to amplify considerably after the fact - example - low volume quiet song in the middle of a loud show - you were running levels conservatively to keep from peaking on the loud pats and now have to amplify (add gain) to the quiet tune - if you add 40 dB you amplify that self noise by the same amount which could become noticeable if the program material is quiet.
No. Amplification afterwards does not change the relationship between the self-noise of the mics, the noise of the rest of your signal chain (preamp, recorder, ect), and the noise-floor of the environment in which the recording was made. You are simply amplifying everything together as you mention - raising the noise floor of the recording by the same amount as the music. The noise-floor of the recording is still going to be dominated by the noise-floor of the environment in which you are recording in almost all cases, even during quiet songs. Yes you may suddenly notice noise during those quieter parts after amplification, but it's just that during those quiet songs you can much more easily hear the noise of the room, which will still be louder than the self-noise of the microphones in almost all live music environments.