The nontrivial aspects of electron structure in lanthanide complexes, considering ligand field (LF) and exchange coupling effects, have been investigated by means of density functional theory (DFT) calculations, taking as a prototypic case study a series of binuclear complexes [LCu(O2COMe)Ln(thd)(2)], where L2- = N,N'-2,2-dimethyl-propylene-di(3-methoxy-salicylidene-iminato) and Ln = Tb, Lu, and Gd. Particular attention has been devoted to the Cu-Tb complex, which shows a quasi-degenerate nonrelativistic ground state. Challenging the limits of density functional theory (DFT), we devised a practical route to obtain different convergent solutions, permuting the starting guess orbitals in a manner resembling the run of the beta electron formally originating from the f(8) configuration of the Tb(III) over seven molecular orbitals (MOs) with predominant f-type character. Although the obtained states cannot be claimed as the DFT computed split of the F-7 multiplet, the results are yet interesting numeric experiments, relevant for the ligand field effects. We also performed broken symmetry (BS) DFT estimation of exchange coupling in the Cu-Gd system, using different settings, with Gaussian-type and plane-wave bases, finding a good match with the coupling parameter from experimental data. We also caught BS-type states for each of the mentioned series of different states emulated for the Cu-Tb complex, finding almost equal exchange coupling parameters throughout the seven LF-like configurations, the magnitude of the J parameter being comparable with those of the Cu-Gd system.