Modeling COVID-19 cases in Africa is crucial for developing effective public health strategies, allocating resources efficiently, and mitigating the impact of the pandemic on vulnerable populations. A recent paper by the first author provided an extreme value analysis of daily new cases of COVID-19 from sixteen countries in west Africa. In this paper, we broaden our analysis to encompass data spanning all fifty four African nations over a period of forty four months. We identified extreme values as the monthly maximums of daily new cases. Utilizing the generalized extreme value distribution, we fitted the data, allowing two of its three parameters to vary linearly or quadratically in relation to the month number. Twenty six countries demonstrated significant downward trends in monthly maximums. Two countries demonstrated significant upward trends in monthly maximums. Nineteen countries demonstrated significant quadratic trends where monthly maximums initially increased before decreasing. The sharpest and weakest of the downward trends with respect to location were for Mali and Liberia, respectively. The sharpest and weakest of the downward trends with respect to scale were for Egypt and Libya, respectively. Recommendations are given for each country. We evaluated the adequacy of fits through probability plots and the Kolmogorov-Smirnov test. Subsequently, the fitted models were employed to determine quantiles of the monthly maximum of new cases, as well as their limits extrapolated to infinite month numbers.