Graphene is different from most optical materials in that it is a thin material layer with a thickness as small as one atom. Graphene layers can be incorporated into optical simulations using either a surface conductivity material model or a volumetric permittivity material model; however, introducing graphene through a volumetric permittivity is computationally inefficient because it requires very fine discretization grids. We have recently developed a more efficient approach that enables the use of comparatively coarse grids by formulating a discretization of Maxwell’s equations (in the time or frequency domains) that combines a surface conductivity description of graphene layers with a volumetric permittivity description of other optical materials. This approach includes the full dispersion characteristics of graphene as specified by the Kubo formula. This paper demonstrates how the combined material description approach can be used to efficiently model state-of-the-art devices that take advantage of the energy confinement provided by surface plasmons. We show how to efficiently model TE and TM polarized surface plasmons, a surface plasmon waveguide switch, and an electro-optical modulator. This last example also includes electrical simulations of graphene and demonstrates how both optical and electrical simulations can be combined to produce a complete model of a graphene based device. For each example, we compare with previously published results, including experimental results.