1) Your Tm are very high, probably you took primers that were too long.
2) Concerning the differences in Tm, when I design the primers, I try to have a number of nucleotides so as to have the same Tm for the 2 primers even if the 2 primers will not necessarily have the same size but will have the same Tm like that I will not have a problem when I calculate the hybridization temperature.
Sofiane Benyamina perfectly addressed the issue. Apart from what she has stated I would like to add that in general, +/_ 3 degree is safe to use but that largely depends on the GC content of the primer.
You can only calculate a Tm for the part of the primer that anneals to the dna template so if your primers include restriction enzyme sites or overlap sequences then your annealing temperatures may be much lower than the temperatures that you have tested
Paul Rutland , I did try as you have mentioned and was unable to get an amplification with phusion enzyme. However, Taq polymerase + DMSO, is giving me a smear. I tried a gradient with Taq considering -5 to +5 Ta (only binding region) of the gene.
A smear may be caused by only one primer annealing and producing single stranded dna. Check that your primers have comlete homology with your sequence ( no SNPs in that part of the sequence) and try lower annealing temperatures and if that fails then design new primers which will cost much less than your time.
Given that the primers are so long there may be secondary structure inter or intra primer problems so borrow some betaine ( very cheap any grade) and run the pcr at a final concentration of 1Molar betaine. If anyone in your lab has any high GC buffer that came with their polymerase that should contain both dmso and betaine and may be worth a try