Strong solvents help. Using dichloromethane works better than acetone, and chloroform works better than dichloromethane. However, there is no way to fully remove the layer post transfer by dissolution alone. A forming gas (Ar/H2) anneal helps in decomposing the PMMA into monomers. Some PMMA strands will remain. It depends on how you cross-linked/cured the PMMA.
Your best bet is to use an alternative polymer for support. We have some forthcoming results on that front which we are going to put into a preprint. Once it is submitted, I would be happy to share it with you.
If you are still transferring using PMMA, then acetic acid is a better choice than acetone. It is very effective in cleaning PMMA. There is no 100% cleaning in solvents. If you clean graphene in acetone then it requires extensive heating in H2/Ar or vacuum. However, using acetic acid, annealing time is reduced to a great extend (may be not required).
At the end, I got my answer from recently published article
Optimized graphene transfer: Influence of polymethylmethacrylate (PMMA) layer concentration and baking time on graphene final performance by Barina et al.
Carbon 84(2015), 82–90
doi:10.1016/j.carbon.2014.11.040
Really nice work. This article will help a lot to researcher like me who are suffering to get PMMA free CVD graphene on desired substrates.