I am not sure I understood your question. I will try to rewrite your question below with my own words and you will let me know if I udnerstood your question correctly.
Description of the genetic of interest
There is only one locus for coat color. Individuals are diploid. At this locus, there exists 4 alleles in the population.
BR individuals are both black and red (as
R are codominant). Another individual is
YW and its phenotype is yellow and white (as
Y are codominant)
What phenotype do
There is no way to know! Relationships between alleles, say
B in determining a given phenotype cannot be extrapolated from relationship
Y have with other alleles. A relationship between two alleles in determining a phenotype is specific to a pair of alleles.
It is for example theoretically possible that
A is dominant over
B is dominant over
C is dominant over
In your case, as you only have codominant relationship, it is tempting to expect that
BY would be both black and yellow but there is really no way to be sure of that from the "Description of the genetic of interest"
You have now changed your post. I don't really understand the new question either. You seem to imply that two alleles may be both dominant. This is impossible. They can be co-dominant.
I think your issue is that you are trying to associate a property of relationship between alleles as a property of an allele itself. Keep in mind, that saying that an allele is dominant only makes sense in a particular context that is in relationship to another allele who will therefore necessarily be recessive.