84 dB would have a lot more noise in the measurements so doesn't make sense to do that. I can however light up the two hidden curves in the frequency response measurements:
View attachment 149310
Visually they look pretty identical. But I took them to Excel and here is the difference (taking out the 10 dB nominal level shift):
View attachment 149311
Hard to know how much of this is due to the headphone, the measurement mic, or general measurement variation. The noise on the right is clearly measurement issue.
I also did the same for 94 vs 114:
View attachment 149312
A bit more pronounced now but again, hard to disentangle from other "nuisance" variables. Also, most of the response is down by about 1 dB so you could just turn up the volume by that amount.