Who says that you couldn't get even better results with a different approach? Your room is obviously well damped at low frequencies. Other's might not have that luxury.
But that's not the point. Gain/level-matching in itself is not a optimization technique that yields smooth frequency response or low modal ringing. How could it?
P.S. Can't read your spectrogram. How long is the impulse window? Can you export impulse responses from your software? Would like to have a look at it.