In typical present-day atmospheric climate models, separate parameterizations are used to represent boundary layer turbulence, stratiform macrophysics, shallow convection, and deep convection. However, the use of separate schemes for separate regimes has several drawbacks. For instance, the use of separate schemes classifies clouds into artificial categories whose boundaries are sometimes violated by nature's continuum. Furthermore, the use of separate schemes makes it difficult to ensure consistency of assumptions across those schemes. Instead, we propose to simulate turbulence and all cloud types in the atmosphere with a single equation set. The equation set consists of Reynolds averages of various moments of the Navier-Stokes and advection-diffusion equations. The equations are closed in part by an assumption about the shape of the distribution of subgrid variability. A key to achieving the requisite generality is the construction of an accurate and flexible subgrid distribution of hydrometeors. We perform single-column simulations of 5 cloud cases: 3 deep convective cases, 1 shallow convective case, and 1 stratocumulus cloud case. The single-column simulations are compared to reference simulations performed by a 3D high-resolution model. At high vertical resolution and short time steps, the single-column results are satisfactory. Remaining challenges include reducing computational cost and troubleshooting the full equation set in a global model.