This paper describes a comparison of the SIRS model, built on the basis of cellular automata, and the real statistics of daily mortality in the course of the COVID-19 pandemic in different countries. The SIRS model is used with different parameters of infectability and loss of susceptibility. The model calculates the cross-correlation of mortality dynamics with the real dynamics of mortality in different countries. A high correlation (more than 0.5) of mortality dynamics in all countries with the dynamics given by the model at the selected parameters is shown. The research results allow us to fix the model parameters that can further perform forecasting. The advantage of the SIRS model based on cellular automata is the simplicity and clarity of a small number of parameters, and the ability to change them in accordance with epidemiological data. The model demonstrates the fact that the daily mortality rate is eventually reduced to zero, although under different parameters (primarily the probability of loss of immunity), the system can survive several waves of infection. Keywords: COVID-19, pandemic, modeling, SIRS, cross-correlation, cellular automata, datasets Distributed in Computational Biology eJournal Vol 4, Issue 41, August 26, 2020. Distributed in Human Health & Disease eJournal Vol 4, Issue 115, August 20, 2020. Distributed in Coronavirus & Infectious Disease Research eJournal Vol 1, Issue 116, September 02, 2020. Distributed in Infectious Diseases eJournal Vol 2, Issue 98, August 17, 2020.