The present paper considers a set of problems of statistical modelling for the transport of solar radiation in stochastic natural media as applied to the problems of remote sensing of the ocean and aerosol and cloudy atmosphere as well as to the problems of constructing the numerical models of solar radiation fields in clouds and cloudy atmosphere. The basis of this consideration is the statistical appro "ach when the radiation transfer is described by the equation, in which some of the parameters are random functions. A set of new Monte Carlo algorithms and programs has been provided for the purpose.This set includes algorithms for the simulation of homogeneous stationary random fields of continuous and broken cloudiness, the optic parameters of the atmosphere and the ruffled sea surface as well as the simulating the radiation transfer process in stochastic media. A special attention has been paid to solving the problem of optimization of Monte Carlo algorithms.