Improving understanding of the two-way interactions between clouds and large-scale atmospheric circulations requires modeling set-ups that can resolve cloud-scale processes, while also including representations of the circulations themselves. In this study, we investigate the potential for mock-Walker simulations to help untangle these interactions by assessing their ability to reproduce the observed climate over the equatorial Pacific. Mock-Walker simulations with realistic zonal sea-surface temperature (SST) gradients show qualitative similarities with reanalysis and satellite data, though notable differences include (1) the presence of double-celled overturning circulations, (2) extreme upper tropospheric dryness over the cold pools, and (3) substantially weaker longwave cloud radiative effects. The double-cell circulations are part of a transition from single to double cells as mean SST is increased, with the transition occurring near present day temperatures. The circulation changes dominate the response of mock-Walker simulations to warming, though their effects are smaller for relatively weak zonal SST gradients. Mock-Walker simulations also exhibit a wide range of climate sensitivities, due to cloud feedbacks that are strongly negative for larger SST gradients and strongly positive for weaker SST gradients. Finally, we show that radiative-subsidence balance can be used to explain the development of the double cells, but are unable to further explain the dynamics of the transition given the complex vertical profiles of stability and atmospheric radiative cooling in these simulations. Since Earth’s present-day climate is close to our simulated transition to a double-celled circulation, these dynamics merit further investigation.