Fuzzy c-regression algorithms such as FcRM (fuzzy c-regression models), which use calculus-based optimization methods, suffer from several drawbacks: they are very sensitive to the presence of noise. Moreover, the memberships are relative numbers. This can be a serious problem in situations where one wishes to generate membership functions from training data. This paper examines how reject options can be used in performing switching regression models. Two types of reject have been included: (1) the ambiguity reject which concerns the data points which fit several models equally well; (2) the distance or error reject dealing with patterns that are far away from all the clusters. To compute these rejects, we use an extension of the Fc+2M algorithm objective function. This algorithm is called the fuzzy c+2-regression model (Fc+2RM).