I primarily do modeling, and I am first in line to tell anyone who is willing to listen that the very first thing you should do to learn to be a good modeler is

I primarily do modeling, and I am first in line to tell anyone who is willing to listen that the very first thing you should do to learn to be a good modeler is spend a year at the bench first in the problem area you want to model. Wet lab is so informationally dense in terms of personal knowledge that it's not even funny.