I've put an update on the HAM7 ISI performance into SEI log 2297, here's a quick summary:

The symmetrization update for the vertical GS-13s on HAM7 has improved the X/RY motion, but not really changed the Y/RX motion.

The coupling at the microseism from GS-13 Z (vertical table motion) to GS-13-RY is about 0.05 rad/meter, and the couping to GS-13-RX is about 0.1 rad/meter. The first set of pdfs show these couplings. These are passive measurements, the results are only valid around the microseism (roughly 120-300 mHz)

The residuals now can be seen in the second set of pdfs - There are for vertical ground BLRMS of about 100 nm/sec. The residual X and Y should scale linearly with the ground Z motion at the microseism. Note that the residuals are plotted as band-limited velocity around the microseism. See the SEI log for more details of how I do this.

The residuals are easy to see. I expect that it will be very difficult to make this any better than the 0.05 coupling to RY.
We'll try, of course, but Jim has already put in significant time, and HAM7 has very good vertical CPS sensors which really help this process.

The signal in the tilt loops can be mostly explained by sensor noise and the remaining vertical coupling.

The way the sensor noise appears in-loop is worked out in T2300378