You don get 5' phosphates with PFU. what you can do is take equi molar concentrations of the primer and treat with PNK to phosphorylate it and then set up your PCR. The amplicon will have 5' phosphates and you can repurify this and directly go ahead with Blunt clonings in pUC19 like vector...
The source of 5' ends in the amplified product is from your primers, not from the Pfu polymerase activity. I assume you are using plain unmodified primers, so your PCR product will not have 5' phosphates. You can either PNK the primers, as suggested by Sathya, or PNK the product, heat inactivate and proceed with ligation - PNK and DNA ligase use the same buffer, so there is no need to purify in between. In general, PNK is more efficient on double stranded DNA (product) than single stranded (primers), but either approach is good enough for cloning.
As mentioned above, your PCR products don't have 5'Phosphate. Two choices - you could order Phosphorylated primer or you could use PNK to put phosphate at the end (use some PEG( (10-20%)) to improve on phosphorylation).
If you just want to join two PCR products you could design primers to overlap (where you andt to join -10-12 bases) and after first PCR you could join these in second PCR
As someone mentioned in the tread for cloning:
I am not sure you will need to have phosphorylated ends on your pcr product, since your vector can provide these. You might ligate in presence of the restriction enzyme that gives blunt end from the vector, if it is absent in your insert. In the old times that was called "forced cloning". Blue/withe screening will give you plenty of positive clones.
Purified PCR products generated by Pfu are suitable for use in blunt-end ligations. They are a bit more tedious to ligate than sticky-end PCR products so carry out the ligation reaction at optimum temp for 4-18 hours.