Typically this is how you would be looking at your public facing services. How many different public ip addresses do you need and for what services then add one for the PAN. You request then from your ISP the appropriate subnet sized for that need, in your case looks like you will need either a /29 or /28. This gets delivered on the ISP device facing your PAN. You use one of these addresses on the PAN. This is now your untrust zone. You now organize your publicly facing resources into risk groups and create the muliple zones and private networks to support them or if the risk is similar they can all go into one DMZ zone. These are the inside interface(s) of your PAN with zone assignments. Now you create your nat rules pointing each ip address from the ISP scope at the matching internal address of the server providing the public service. And write the security policies needed for the inbound and outbound communications for each server.
... View more