I think the first thing to understand about ball pythons genetics is that we use the term co-dominance incorrectly. To date, there is no such thing as a "co-dominant" morph in ball pythons. Period. Everything that we call co-dominant is actually "incomplete dominant" but the other term is used so commonly that we're sort of stuck with it. So you can completely disregard any confusion you have about the true definition of co-dominance not matching ball python genetics. Read the first sentence of the description of Mojaves: http://www.worldofballpythons.com/morphs/mojave/
http://reptilesbreedingenterprise.com/Genetics_3.html
http://www.newenglandreptile.com/ind...dominant-genes
Besides, co-dominance requires two fully dominant alleles expressing themselves at the same locus. Even if we were using the term correctly, Mojave and Cinnamon are found at different loci, so they cannot be described as co-dominant to each other, or even incompletely dominant to each other.
For example:
Lets say the Mojave allele is represented by M, and the normal allele at the same locus is represented by m. Incomplete dominance simply means that the Mm doesn't look exactly like MM or mm. There is a super mojave, a mojave, and a normal... meaning two M's is more phenotypically intense than just one M (hence the term incomplete dominance).
With the Cinnamon trait, the heterozygous cinnamon (Cc) looks different from either the super (CC) or the normal (cc).
Meanwhile, a Mojave Cinnamon would be MmCc, not MC. Mojave is incomplete dominant, and Cinnamon is incomplete dominant, but they are not incomplete dominant to each other. This is called independent assortment and it is crucial to understanding combo morphs.