gPPI different task in different runs

Dear AFNI Experts,

I am using afni to perform gPPI to my fMRI data following the scripts in AFNI_data6.
My experimental design is: in 3 runs, participants perform task1, in another 3 runs, participants perform task2. The stimuli in these runs are same. I want to see whether the task would modulate the activity or connectivity between seed ROI and other regions. When I look at the tutorials I found that usually different task or stimuli were performed in same runs. I wonder whether my design would cause any artifacts? I used errt.allruns to calculate the seed time series and the created PPI regressors look like this :
Do you think there is any problem? Will the motion regressors cause any artifact to this, since the motion regressors for different runs were created separately?
Thanks!

Screen Shot 2019-05-31 at 5.50.10 PM.png