Understanding difference between models via wavefunctions

Hello,

The problem is that the model I am investigating is not well documented. The resonance there is supposed to be Z prime, but it somewhat looks different from the SM Z based on a wavefunction flow. I was unable to pin down a clear reference on HELAS that explains the different numbers in the wavefunctions calls, i.e., FFVX_Y_Z, what XYZ stand for. The difference between the two models boils down to a difference between using FFV2_3_3 (for SM Z) and using FFV2_5_3 (for Z'). The code snippet looks as the following:

ixxxxx(p[perm[0]], mME[0], hel[0], +1, w[0]);
oxxxxx(p[perm[1]], mME[1], hel[1], -1, w[1]);
oxxxxx(p[perm[2]], mME[2], hel[2], +1, w[2]);
ixxxxx(p[perm[3]], mME[3], hel[3], -1, w[3]);
oxxxxx(p[perm[4]], mME[4], hel[4], +1, w[4]);
ixxxxx(p[perm[5]], mME[5], hel[5], -1, w[5]);
FFV1P0_3(w[0], w[1], pars->GC_2, pars->ZERO, pars->ZERO, w[6]);
FFV2_3_3(w[3], w[2], pars->GC_9, pars->GC_10, pars->mdl_ZpM, pars->mdl_ZpW, w[7]);
FFV1_1(w[4], w[6], pars->GC_3, pars->mdl_MMU, pars->ZERO, w[8]);
FFV2_5_3(w[3], w[2], pars->GC_29, pars->GC_34, pars->mdl_MZ, pars->mdl_WZ, w[11]);

// Calculate all amplitudes
// Amplitude(s) for diagram number 0
FFV2_3_0(w[5], w[8], w[7], pars->GC_11, pars->GC_12, amp[0]);
FFV2_5_0(w[5], w[8], w[11], pars->GC_29, pars->GC_34, amp[4]);

If I match Z and Z' couplings, masses, and widths, amplitudes are still different: an unexpected result based on a model description. However, I clearly see that different calls are being used FFV2_3_3 vs. FFV2_5_3.

Could somebody explain what is going on?
Where can I read myself for future reference on a precise wavefunction building in MadGraph5_aMC@NLO?
(HELAS paper from '92 lacks content here)

Thank you so much!

Best,
Aurelijus