The aim of this work is to generalize some existing results dealing with the unbiased H∞ filtering for dynamical systems. In particular, we consider reduced-order filtering for a large class of uncertain linear neutral systems subject to unknown inputs and external disturbances. Our purpose consists in ensuring both: the total decoupling with respect to the unknown inputs and minimization of the external disturbances effects using the H∞ criterion. Both cases of known and unknown delay are considered. Sufficient conditions for the existence and the stability of the proposed filters are derived. Finally, we succeed in transforming the filter design problem into the feasibility of some reduced-order linear matrix inequalities. Theoretical and numerical comparisons with others important works are provided to show the effectiveness of the obtained results.