How to remove unique columns of a 3D matrix

I am trying to build a somewhat large 3D matrix. I'm running into a dimension mismatch at the end of my code when I am removing junk columns and concatenating others.

Here's the code that I'm having trouble with: Some other information, the size of stims is 6 x 189 x 96

      blank  = [0; 0; 0; 0; numBlank; meanBlank];
      white1 = [30; 0; yloc(1,1); xloc(1,1); numWhite1; meanWhite1];
      white2 = [30; 0; yloc(1,2); xloc(1,2); numWhite2; meanWhite2];
      white3 = [30; 0; yloc(1,3); xloc(1,3); numWhite3; meanWhite3];
      black1 = [60; 0; yloc(1,1); xloc(1,1); numBlack1; meanBlack1];
      black2 = [60; 0; yloc(1,2); xloc(1,2); numBlack2; meanBlack2];
      black3 = [60; 0; yloc(1,3); xloc(1,3); numBlack3; meanBlack3];

      % remove the columns that were combined above
      delBlank = stims(1,:,ch) == 0;
      delFlash = stims(1,:,ch) >= 30;

      del = delBlank | delFlash;

      stimStrip = stims; 
      stimStrip(:,del,:) = [];
      % concatente the blank and flash vectors to the end of the matrix
      stimStrip(:,:,ch) = horzcat(stimStrip(:,:,ch),white1,white2,white3,black1,black2,black3,blank);
      % add to data structure to save later
      data.stims(:,:,ch) = stimStrip;