In a complementary article [1], we exploited algebraic properties of Maxwell's equations and fundamental principles such as electromagnetic reciprocity and passivity, to derive fundamental limits to radiative heat transfer applicable in near-through far-field regimes. The limits depend on the choice of material susceptibilities and bounding surfaces enclosing arbitrarily shaped objects. In this article, we apply these bounds to two different geometric configurations of interest, namely dipolar particles or extended structures of infinite area in the near field of one another, and compare these predictions to prior limits. We find that while near-field radiative heat transfer between dipolar particles can saturate purely geometric "Landauer" limits, bounds on extended structures cannot, instead growing much more slowly with respect to a material response figure of merit, an "inverse resistivity" for metals, due to the deleterious effects of multiple scattering; nanostructuring is unable to overcome these limits, which can be practically reached by planar media at the surface polariton condition.Radiative heat transfer (RHT) between two bodies may be written as a frequency integral of the formwhere Π(ω, T ) is the Planck function (and it has been assumed, without loss of generality, that T B > T A so P > 0), and Φ(ω) a dimensionless spectrum of energy transfer. RHT between two objects sufficiently separated in space follows the Planck blackbody law, but in the near-field where separations are smaller than the characteristic thermal wavelength of radiation, contributions to RHT from evanescent modes will dominate, allowing Φ(ω) to exceed the far-field blackbody limits by orders of magnitude. Moreover, because the Planck function decays exponentially with frequency, judicious choice of materials and nanostructured geometries can shift resonances in Φ to lower (especially infrared) frequencies, allowing observation of even larger integrated RHT powers [2-5]. However, after accounting for the effects of such frequency shifts, the degree to which the spectrum Φ at a given frequency can be enhanced remains an open question. The inability of trial-and-error explorations and optimization procedures [6, 7] to saturate prior bounds on Φ based on modal analyses [8][9][10][11] or energy conservation [12] suggests that these prior bounds may be too loose.In a complementary article [1], we derived new bounds that simultaneously account for material and geometric constraints as well as multiple scattering effects. These bounds, valid from the near-through far-field regimes, incorporate the dependence of the optimal modal response of each object on the other while simultaneously being constrained by passivity considerations in isolation. They depend on a general material response factor ("inverse resistivity" for metals) [12],without making explicit reference to specific frequencies or dispersion models, and are domain monotonic, increasing with object volumes independently of their shapes. Consequently, our bounds are applicable ...