Clouds are important: they strongly influence Earth's radiation budget and thereby affect Earth's temperature, precipitation, and atmospheric circulation. Clouds are also intriguingly complicated: they span a large range of spatial scales and involve many tightly coupled processes, including dynamics, thermodynamics, radiation, microphysics, and aerosol chemistry. Such complexity makes it very difficult to unravel the essential mechanisms of clouds and to predict their responses and feedbacks in different climate regimes. Tan uses a hierarchy of models that span across various scales, including idealized general circulation models (GCM), single column models (SCM), and large-scale simulations (LES). Idealized experiments are designed for these models for simplification to the extent possible that the fundamental physical constraints remain valid. Understanding on the essential mechanisms of clouds are thereby developed, and simple analytical models that capture the key processes can be constructed accordingly. This new knowledge can be applied in the actual climate change context, which helps us to reduce the theoretical and modeling uncertainty on cloud responses and feedbacks.