JPJP: you correctly assess the issue, but you somewhat fail to highlight the risk wherein (if you'll please excuse me saying so?!) just because an IAN derived approach 'looks' like an ILS, one need always remember that is is not an ILS signal derived from and / or emanating from a position that is know & fixed along the runway, i.e. an IAN approach is derived from multiple inputs that are 'integrated' within your aircraft's FMC to generate approach guidance that seemingly reflects what one would see were one following an ILS (but it is NOT an ILS)... and if ever proof of that were needed, try flying an IAN approach when it is either very cold (i.e. well below ISA, indeed Boeing themselves tell you not to do IAN when it's v.cold) or conversely when it is very hot (i.e. well above ISA, when you'll likely break out at minimums with 3, or maybe even 4, 'whites').... and even v.10.8.A of the FMC does not (empirically) deal with significant ISA temperature errors in this respect (of IAN specifically; though one can somewhat modify this issue wrt to VNAV derived approaches).
The point of flying a non-precision approach (VOR) via IAN is to make the process as consistent & simple as possible (much as per the ILS procedure) but always with the caveat that one can follow the IAN guidance only down as far as the MDA (i.e. the last point by which one must 'visually' acquire the runway) and after that point (i.e. below the MDA) one must fly 'visually' to the runway and should not rely upon the IAN's Flight Director guidance (no matter how good a job it might seem to be doing... always remembering that IAN is not an ILS).
The point about turning off (then back on) of the Flight Directors, when following vertical guidance that is non-precision, is that that the vertical guidance is precisely that, i.e. NON-precision, being that it is generated onboard your aircraft and therein is subject to all manner of errors, whereas an ILS signal which is projected from the ground is subject to very few errors (in the most instances).