QUESTIONS
01 How does eBox decide which of the ranges to allocate a lease from?
02 What is the purpose of eBox being able to create multiple ranges within a larger available range?
I think there's some confusion here. When eBox's DHCP module is first enabled, the entire address range is "unallocated." You could assign static IPs from the entire range. The purpose of DHCP is to allow at least some client devices to have a dynamic addresses allocated as needed. So out of the entire range of available addresses a DHCP range or dynamically available range of addresses is set aside. Any address outside the DHCP/dynamic range may still be assigned statically.
So with that, the questions, but in reverse order:
02 I'm not sure what the purpose is, unless it's meant as a flexibility feature (perhaps a single eBox can allocate addresses from different subnets?). Having two or more DHCP ranges seems to me to be unnecessary in most cases. A single block of addresses big enough to accommodate the number of possible devices works well.
01 Again, I don't know how this might work, because I'm not sure what purpose the capability serves. But for the task of replacing a single router, you need to assign only one DHCP/dynamic address range. All addresses outside that range are by definition unallocated and can be assigned statically as needed.
EDIT: A case where you might want to have some addresses available for static assignment: A wireless access point. Assigning the access point a static address means you always know where to find it. Another might be a network device you want to give special treatment, like an Xbox. So while it's possible to put an entire subnet into a DHCP range, if it's not necessary to do so, it's better to have some unallocated (from the DHCP server's point of view) addresses available for these types of purposes.