While there has been much success in modeling the linear and nonlinear rheology of monodisperse entangled linear polymers, progress in the constitutive modeling of polymeric materials continues to lag behind the needs of industry. Industrially sourced polymers are typically polydisperse (comprising a broad distribution of molecular weights), making their rheology more suitable for processing but also more difficult to predict. To date, there are no molecular-based constitutive models that are practically suitable for describing industrially relevant polymers in industrially relevant flows. In this article, we extend but strongly simplify the model of Read et al. [J. Rheol. 56, 823-873 (2012)], which is able to predict the linear and nonlinear rheology of bidisperse blends but is prohibitively complex for industrial use. We propose a pair of simplified tube models for polydisperse melts of entangled linear polymers that combine the success of the double reptation approximation [des Cloizeaux, Europhys. Lett. 5, 437-442 (1988)] in the linear regime with the success of the Rolie-Poly constitutive equation [Likhtman et al., J. Non Newtonian Fluid Mech. 114, 1-12 (2003)] in the nonlinear regime. We first review the key concepts of the double reptation approximation and the original (monodisperse) Rolie-Poly constitutive model. Subsequently, we provide the details of our approximate models for the particular case of a bidisperse blend and show that these models naturally identify the effects from couplings between constraint release and chain retraction (i.e., the so-called "enhanced stretch relaxation time"). Finally, we generalize to a multicomponent (polydisperse) model, based on the same underlying principles. Along the way, we also show that both of our models are in qualitative, and largely quantitative, agreement with experimental data for bidisperse and polydisperse melts of entangled linear polymers.