We have performed kinetic Monte Carlo simulations of benzene diffusion in Na-Y at finite loadings for various temperatures to test the analytical theory presented in Paper I, immediately preceding this paper. Our theory and simulations assume that benzene molecules jump among S-II and W sites, located near Na+ ions in 6-rings and in 12-ring windows, respectively. The theory exploits the fact that supercages are identical on average, yielding D-theta = 1/6k(theta)a(theta)(2) = kappa a(theta)(2)/6[tau(1)][1 + K-eq(1-->2)], where k(theta) is the cage-to-cage rate coefficient, K-eq(1-->2) is the W-->S-II equilibrium coefficient, [tau(1)] is the mean W site residence time, and kappa is the transmission coefficient for cage-to-cage motion. The simulations use fundamental rate coefficients calculated at infinite dilution for consistency with the theory in Paper I. Our theory for k(theta), K-eq(1-->2) and [tau(1)] agrees quantitatively with simulation for various temperatures and loadings. The simulated transmission coefficient is nearly 1/2 for all but the highest loadings, qualitatively validating our mean field approximation. Comparison between our theory and experimental data shows excellent qualitative agreement with tracer zero-length column data, but also shows qualitative disagreement with, both pulsed field gradient NMR and frequency response data. (C) 1997 American Institute of Physics.