The representations of clouds in climate model MRI-ESM2 (Yukimoto et al. 2019, JMSJ) used in CMIP6 simulations are significantly improved from the previous version MRI-CGCM3 used in CMIP5 simulations. The score of the spatial pattern of 2m temperature for MRI-ESM2 is better than any of the 47 CMIP6 models. The score for radiative fluxes at the top of the atmosphere for MRI-ESM2 is ranked in the top seven models among these CMIP6 models. We will introduce comprehensively various modifications related to clouds, which contribute to the dramatically improved cloud representation, and their main impacts (Kawai et al. 2019, GMD). The modifications cover various schemes and processes including the cloud scheme, turbulence scheme, cloud microphysics processes, the interaction between cloud and convection schemes, resolution issues, cloud radiation processes, aerosol properties, and numerics. In addition, we would like to mention what data we used during the model development, how we evaluated our model, how useful the intercomparison studies were, and what data we need for further model development.