The jellium surface energy has long been of interest to the Density Functional Theory (DFT) community, since the jellium surface is the simplest example of a strongly inhomogeneous density. Over three decades, various methods have been employed to investigate this problem, and the answers differ significantly from one method to another. In recent years, the appearance of the Quantum Monte Carlo surface energies, which are significantly higher than those of DFT, has raised a serious concern for density functional theorists, since the results from Quantum Monte Carlo are usually considered 'exact'. In this work, we investigate the problem with: (1) a long-range correction to the generalized gradient approximation (GGA); and (2) a GGA short-range correction to the random phase approximation (RPA). The results from these two approaches and those from another advanced density functional, the PKZB meta-GGA, agree among themselves within one percent. Our consistent results strongly support the older and less-sophisticated DFT estimates of jellium surface energies, and not the Quantum Monte Carlo values Our GGA's for the exchange-correlation energy within and beyond RPA are constructed by a real-space cutoff of the spurious long-range contribution to the second-order gradient expansion for the exchange-correlation hole around an electron. In a loosely related study, we show that the sixth-order gradient expansion for the kinetic energy, which diverges for surfaces and finite systems, provides useful accuracy for bulk solid