Aliovalent cation substitutions in MO2 drive the formation of oxygen vacancies as a charge compensation mechanism. There is evidence for systematic ordering of vacancies in non-stoichiometric MO2-x systems (M = Pr, Ce, and Tb) leading to the formation of several intermediate fluorite-related phases. In pseudo-binary oxide systems of the family Ln2O3:MO2 with Ln3+= La, Nd, Gd, ..., oxygen vacancies are also formed to maintain charge neutrality. The existing literature about these mixed systems reports a perfect solubility of the aliovalent cations and postulates a random distribution of the anion vacancies suggesting an ideal solid-solution behaviour and a monotonic increase of their concentration over the whole range of compositions: a smooth change between the α-type structure (fluorite) and the A-type or C-type Ln2O3 structure is usually assumed. Nevertheless, the average lattice parameter of this description seems to display some departure from the expected linear behaviour. Indeed, by analogy with the intermediate phases of the PrO2-x system, a cornucopia of phases would be expected to form also in a pseudo-binary system. However, the cation mobility and their capacity to rearrange at various temperatures decrease rapidly with temperature and this changes the extent of the problem as kinetic considerations become a relevant feature. Chemical homogeneity and charge ordering of the cation sublattice of these systems would require high temperature and prolonged annealing that are only effective just above the charge-ordering transition temperature. Moreover, this annealing should occur over a time span dictated by the chemical diffusion coefficient: achieving these two conditions (ordering occurring only when the cation inter-diffusion is sluggish) can be practically impossible in samples produced by powder metallurgy techniques. In order to achieve chemical homogeneity without relying on cation diffusion and, at the same time to minimize non-equilibrium processes, various compositions of NdxCe1-xO2-x/2 were fabricated at low temperature as nanocrystalline powders using a freeze-drying method. The structures of these compounds were studied by x-ray diffraction and selected area electron diffraction. In these homogeneous systems, we observe a departure from the random oxygen-deficient fluorite structural model. A set of structures derived from the fluorite were used to model the effect of the vacancy order upon the diffracted intensities. While the average fluorite structure remains present at moderate doping levels, a phase change occurs in the region 0.36 < Ndx < 0.48, marked by the appearance of low-intensity superstructure peaks in the diffraction patterns. The powder diffraction data is best described by a monoclinic Ln6O11-type structure (SG P21/c), often referred to as “β phase” in PrO2-x. These findings question the equilibrium character of the current phase diagrams and require a possible critical reassessment of several of these pseudo-binary systems. This work was funded by EPSRC through the Next Generation Nuclear Centre for Doctoral Training (NGN-CDT) programme.