This paper expounds that the element part belongs to the virtual fuzziness presented by the fuzzy set, why not be the object of the fuzziness research, why the fuzzy set theory taking the virtual fuzziness as the research object is not successful and the negative impact on the normal fuzziness research. It is pointed out that the fuzziness to be studied has incomplete and imperfect information, and the "truth value" of the target value can not be known that only the uncertainty generated by the approximation of the truth value can be determined. In the fuzziness study, a supporting logic should be established to help solve the problem of non-repeatability of calculation, non-unique calculation results and not "relative optimal approximation" in the fuzziness study. The most qualitative point is to solve the problem of "algorithm realization" of knowledge transformation under the relevant relationship, which is the key point of whether the fuzziness research can succeed and whether the fuzziness research can become the branch of mathematics research like the randomness research.