Since pycnodysostosis is caused by mutations that cause deficiency in Cathepsin K, it makes sense to look at animal models for the latter; there's a fairly recent study Dauth, 2011...
Cathepsin K deficiency in mice induces structural and metabolic changes in the central nervous system that are associated with learning and memory deficits. [...]
Cathepsin K is a mammalian cysteine peptidase that is sorted to endo-lysosomes and secreted into the extracellular space by certain cell types . It is abundant in osteoclasts and has a prominent role in bone remodeling due to its collagenolytic activity . Indeed, cathepsin K deficiency has been linked to bone disorders such as pycnodysostosis and osteopetrosis [3,4], while the excessive activity of this enzyme may lead to osteoporosis , making it a likely target for rational drug design [6-8].
Interestingly, pycnodysostosis patients suffering from lifelong absence of cathepsin K activity have been occasionally reported to exhibit, in addition to the abnormalities in bone development, alterations in the CNS such as hyperplasia of the pituitary, demyelination of the cerebrum, and imbalances between brain growth, vascular supply, and cerebrospinal fluid pressure, with complications including sensorineural deafness and frontal porencephalic cysts. 
Unfortunately the trail gets pretty difficult from there because the paper cited for the last para, while mentioning "mental retardation" as known symptom, block-cites some 5 papers for that; I checked out one of them (in JAMA) but that one doesn't mention anything mental, so... who knows which of the other case reports might have some cognitive details, if any.
I guess the incidence of cognitive issues might depend on the specific mutation, since GARD says that only 5-29% of PSYD actually have these cognitive abnormalities.
Research on animal models of some genetic diseases usually starts with a substantial mutation, i.e. something that's nearly sure to cause the desired disease effect. Later on milder variants are studies. (For example see how the animal models of Huntington's advanced) With pycnodysostosis, at least as far neuronal/behavioral effects go, it seems we're at the fist stage (i.e. they published the publishable "sure thing").
The highly cited paper of Gelb, 1996, which catalogued the (few) known mutations for pycnodysostosis alas does not correlate them with symptoms beyond the mere diagnosis of pycnodysostosis. In fact, this paper doesn't even mention any mental/cognitive effects in the patients studied... which were alas only two or three families.