Good to know! So my impression wasn't wrong.I am not an expert on acoustics, and am an electronics guy so let me put this it in a way that I understood myself - if I misrepresnt it I am sure more knowledable people can correct me.
There is just a relevant point you did not take into account. I did not say that two IEMs that measure the same sound the same for EVERYBODY, but for a single person. If you combine the nonlinear output impedance e(x) with the nonlinear input impedance of a specific ear canal g(x) and the output impedance of two IEMs is identical (since it was measured) e(x) = f(x), doing a convolution with the input impedance will result in the same f(g(x)) = e(g(x).An IEM is a device that has an output impedance that varies with frequency. An ear canal also has an input impedance that varies with frequency. When you couple these devices, the final FR depends on the interactions between the two variable impedance systems and is not a linear function of either. That is why the same IEM, measures same in some frequencies and different in some others in two different measurement rigs, and that is why it might procude a totall different FR in your ear canal, which has a different impedance/FR curve.