The exciting dark matter (XDM) model was proposed as a mechanism to efficiently convert the kinetic energy (in sufficiently hot environments) of dark matter into e +e - pairs. The standard scenario invokes a doublet of nearly degenerate dark matter (DM) states and a dark force to mediate a large upscattering cross section between the two. For heavy (∼TeV ) DM, the kinetic energy of weakly interacting massive particles in large (galaxy-sized or larger) halos is capable of producing low-energy positrons. For lighter dark matter, this is kinematically impossible, and the unique observable signature becomes an x-ray line, arising from χ χ →χ*χ*, followed by χ*→χ γ . This variant of XDM is distinctive from other DM x-ray scenarios in that its signatures tend to be most present in more massive, hotter environments, such as clusters, rather than nearby dwarfs, and has different dependencies from decaying models. We find that it is capable of explaining the recently reported s-ray line at 3.56 keV. For very long lifetimes of the excited state, primordial decays can explain the signal without the presence of upscattering. Thermal models freeze out as in the normal XDM setup, via annihilations to the light boson ϕ . For suitable masses, the annihilation χ χ →ϕ ϕ followed by ϕ →SM can explain the reported gamma-ray signature from the Galactic center. Direct detection is discussed, including the possibility of explaining DAMA via the "luminous" dark matter approach. Quite generally, the proximity of the 3.56 keV line to the energy scale of DAMA motivates a reexamination of electromagnetic explanations. Other signals, including lepton jets and the modification of cores of dwarf galaxies are also considered.