In a previous paper, an approach has been proposed to solve the constrained reciding horizon optimal problem (CRHOP) for nonlinear discrete plants under certain assumptions [1], the idea being to split the controlactioninto a linearizing control, which forces the plant to behave like a linear reference model, and a standard linear CROHP, which is applied to the linearized plant. This approach has been shown to solve exactly the discrete CROHP for a wide class of nonlinear plants, and offers an alternative view of the problem, in which the feedback linearizing condition is substituted by a model matching requriement. New conditions for the optimality of the solution are thus derived.