As the main ways of the coronavirus (COVID-19) spreading are airand physical contact, actions to mitigate and suppress its spreading must bedeveloped in order to change population behavior, providing efficient controlstrategies. Here, these actions are described as a simple heuristic frameworkto establish public policies. The possible effects of these actions are modeledas a stratified control algorithm applied to an extension of the Susceptible-Infected-Removed (SIR) compartmental model. The model dynamics is an-alyzed and validated with simulations considering parameters obtained fromepidemiological data from Brazil and Uruguay. Model simulations allow theproposition of a three-level control strategy that, as the numerical experimentsshow, presents promising results for the development of strategies to reducethe spread of COVID-19.