I know they are a voltage divider, but it's unnecessary because you can get the same effect by lowering R4. Or just jumper R18 & C5, delete R9 & R19, reduce R10 to 82K and you have the exact same transfer function with fewer parts.
Since you are splitting the Grover in half, now I see why you're doing what you're doing. Consider making Q1 Si, it will make the bias much more stable. Q2 dominates the tone because that's where the distortion is made. Use a medium-low HFE transistor for Q1 or increase R3 to dial-in the bias.
I like the idea of 100nF in parallel with R4, it rolls-off the higher harmonics above 3.4KHz. If that's too dark, then try 47nF.
Because of the feedback thru R3, a FF behaves differently than a Rangemaster. Varying the GAIN will change the high-pass freq because the gain of Q1 & Q2 makes R3 proportionately smaller from an AC point of view.