Is the sky_diffuse_passias model conceptually correct?

[kandersolar]

pvlib.shading.sky_diffuse_passias implements a row-to-row diffuse irradiance reduction model given in Passias & Kallback 1984. The reference uses the term "masking" to refer to row-to-row diffuse shading and claims that "the relative reduction in the diffuse irradiance due to the masking effect" is given by $1 - \cos^2(\psi/2)$, where $\psi$ (the "masking angle") is the angle between horizontal and the ray from the evaluated point on the shaded module to the top of the shading row:

I see that $1 - \cos^2(\psi/2)$ is equivalent to $(1 - \cos(\psi))/2$ and thus can be understood as a case of the B-71 view factor where the obstruction is not somewhere in the middle of the field of view like the above figure suggests but rather at the edge of the field of view as shown in pink here:

So it seems to be accounting for shading from some other obstruction rather than the adjacent row. I also don't understand how it could account for row-row shading without taking the surface tilt into account.

I find myself even more skeptical after comparing the equation with the VF calculations elsewhere in pvlib and not getting the values I expect:

In [1]: from pvlib.bifacial.utils import vf_row_sky_2d  # pvlib v0.10.1
   ...: from pvlib.shading import masking_angle, sky_diffuse_passias
   ...: from pvlib.irradiance import isotropic
   ...: import numpy as np
   ...: 
   ...: gcr = 0.5
   ...: surface_tilt = 30
   ...: x = np.linspace(0, 1, 4)
   ...: 
   ...: psi = masking_angle(surface_tilt, gcr, x)

In [2]: # reference view factor accounting for both ground and row-row "masking"
   ...: print(vf_row_sky_2d(surface_tilt, gcr, x))
[0.79534525 0.86456325 0.9067407  0.9330127 ]

In [3]: # matches value from vf_row_sky_2d, but isn't what the reference intended
   ...: print(1 - sky_diffuse_passias(psi + surface_tilt))
[0.79534525 0.86456325 0.9067407  0.9330127 ]

In [4]: # how it seems sky_diffuse_passias should be used
   ...: print(isotropic(surface_tilt, 1) * (1 - sky_diffuse_passias(psi)))
[0.89336063 0.92071165 0.93081593 0.9330127 ]

Am I misunderstanding how to use the Passias equation? How can it be used correctly?