You could even throw a tone control right after C4.
I wouldn't suggest this, at least not without bread-boarding it first (so you don't have drilled holes and extra pots which may not get used).
First, the added tone controls will be inside a feedback loop, which can make their operation be very counter-intuitive.
Second, the 6L6 is harder to drive than the 6V6, which is at least part of the reason the feedback resistor was raised from 22kΩ (original Princeton) to 68k (Maggie). The feedback loop sets the gain from input to 2nd stage to speaker at ~46. It takes 13.3v peak across the terminals of an 8Ω to make 11w (which seems a reasonable target output power), so you need 0.29v peak at the input to the 2nd stage. 1st stage gain will be around 60 giving sensitivity of 5mV at max volume and guaranteed distortion.
But half-volume is also 1/10th voltage output to 2nd gain stage, so you'll need 50mV at the input jack for max output. You'll get some hair on the notes
unless you're in the center-off flat position of the voicing switch. That leaves the cathode resistor unbypassed, and cuts 1st stage gain to ~30. Max-volume sensitivity drops to ~10mV (still distorts full-volume), while half-volume give full power with 100mV at the input jack (clean unless you have humbuckers or *very* hot single coils).
So the amp's gain is structured right for what it is now. If you add a tone control more-elaborate than the single-knob circuit, insertion loss will the amp seem hard to distort, which is probably counter to why you'd build a SE amp in the first place. More knobs, more adjustment seems like a good thing until you actually use it, when things tend to be set & forget. I think the Maggie has the right mix of stuff for what is intended to be a simpler small amp where you fiddle less and play more.