In this and a previous article (J. Phys. Chem. A 2000, 104, 8244), the range of application for relativistic density functional theory (DFT) is extended to the calculation of nuclear magnetic resonance (NMR) shieldings and chemical shifts in diamagnetic actinide compounds. Two relativistic DFT methods are used, ZORA ("zeroth-order regular approximation") and the quasirelativistic (QR) method. In the given second paper, NMR shieldings and chemical shifts are calculated and discussed for a wide range of compounds. The molecules studied comprise uranyl complexes, [UO2Ln](+/-q); UF6; inorganic UF6 derivatives, UF6-nCln, n = 0-6; and organometallic UF6 derivatives, UF6-n(OCH3)(n), n = 0-5. Uranyl complexes include [UO2F4](2-), [UO2Cl4](2-), [UO2(OH)(4)](2-), [UO2(CO3)(3)](4-), and [UO2(H2O)(5)](2+). For the ligand NMR, moderate (e.g., F-19 NMR chemical shifts in UF6-nCln) to excellent agreement [e,g., 19F chemical shift tensor in UF6 or H-1 NMR in UF6-n(OCH3)(n)] has been found between theory and experiment. The methods have been used to calculate the experimentally unknown U-235 NMR chemical shifts. A large chemical shift range of at least 21000 ppm has been predicted for the U-235 nucleus. ZORA spin-orbit appears to be the most accurate method for predicting actinide metal chemical shifts. Trends in the U-235 NMR chemical shifts of UF6-nLn molecules are analyzed and explained in terms of the calculated electronic structure. It is argued that the energy separation and interaction between occupied and virtual orbitals with f-character are the determining factors.