Using SSL, you get point-to-point encryption between the client and the service. If the service is not the final recipient of the message, but a proxy server that sends this message to another service, you do not have encryption between the two services or you also need to configure this.
WS-Security configured through WS-Policy can give you end-to-end encryption between the client and the final recipient of the message, since you can encrypt the message body. You do not need to configure SSL for each pair of communication objects. Each proxy server can simply forward the message as defined in the header.
However, if you do not need end-to-end guarantees, but a two-point guarantee is enough (this is your scenario, as I understand it), I would say that using SSL is a fair choice.
Another thing to consider is that the WS-Security client and service implementations must be able to communicate. SSL is usually quite mature, but my personal experience is that WS-Security implementations are not. Thus, if you have different WS- * Stacks for client and server, perhaps some hackers and trial errors will find a policy configuration that works for both.