This paper deals with quantitative aspects of the agglomerate model for MCFC cathodes. The behavior of the model is validated and input parameters are obtained for the non-linear steady-state model, by fitting the time-dependent linearized model to a series of ac-impedance spectra of porous LiCoO2 cathodes. The resulting partial-pressure dependences of the exchange-current density are in favor of the peroxycarbonate mechanism and the oxygen mechanism involving oxygen atoms. The applicability of the equivalent model concept as proposed in Part I [1], is confirmed by the obtained parameter values. When determining physical parameters from fit results, the actual agglomerate shape appears to be very important. For performance-optimization studies, the effect of small parameter variations on the polarization resistance has been determined at -100 mV. The results are similar to the ones found at OCV. The ionic conductivity was found to have the largest impact on the polarization resistance, while increasing the rate of mass transport is only useful if this is done for both reactants simultaneously. The total polarisation cannot be written as a simple sum of individual contributions. Only cautious indications can be given for individual contributions of the various polarization sources from limiting cases. Based on analytic results of Part I [1], the sizes of dead cores in agglomerates due to depletion of reactants have been estimated. Only for fairly poor gases a dead core may be observed. The sufficient electrode thickness is predicted to be several times smaller than currently applied in fuel cells.