This paper proposes a model which makes an attempt to incorporate heterogeneous labor and imperfect information in the theory of sequential and circular migration of labor. An attempt has been made in this paper to capture the individual’s migration decision and behavior based upon maximizing expected utility of terminal wealth. Here, the individuals are considered to be heterogeneous in terms of their human capital endowment. The incentive to make the initial move is determined by a solution to a stochastic control problem. The reason for posing migration as a stochastic control problem is that the outcome of the migration is uncertain. The result of this stochastic control problem has been extended to adaptive control processes of a learning sort. Circular migration occurs via a continuous process of evaluation and reevaluation of labor market conditions in different regions. The model presented in this paper provides a mechanism which allows us to identify certain decision rules regarding the process of sequential and circular migration. The model provides a theory which can extensively explain the mechanism of sequential and circular migration under a stochastic framework.