One thing I wonder about the Rhind and Moscow Papyrus problems is that they seem concerned with what I would call operations research problems - ie how much grain would fit in a container of a cylindrical grain silo, a hemispherical basket etc. In this context, due to the errors in construction of the said containers - ie they are not perfectly circular or hemispherical, there is little practical need for high accuracy.
This appears to contradict the otherwise strong desire for accuracy - "It was decreed that each cubit stick should be returned at each full moon to be compared to the Royal Master. Failure to do so was punishable by death."
Why place such a high value on accuracy of measurement if your most advanced mathematics is for problems of grain storage / taxation?