i suggest 2 options

Option 1: if you want a RBDMs, i think 2 tables will be sufficient

Table 1 (Lookup)
-LocationID
-Continent
-country
-region
-City
-Landmark (granularity)

in this Table since its a lookup so i suggest keeping it Denormalized, it will save you alot in your coding specially when joining tables and searching and reporting. and maintenance is easier, then 5 separate tables

the other table is the table to collect the data

Table 2 (Master)
-Tourist
-LocationID
-ExtraFields
-...

so in this table you collect info about tourists and all you have to do is store the locationid, which is your granularity, this way the Locationid, has the Landmark,city,region,country,continent Since the Landmark is definetly in a city and the city is in a country and the country is in a region.. (no need for 5 tables to keep them separate, there will be a big chance of invalid inputs or data if not maintained properly)

Option 2:

i suggest graph database, check this post at SO , about using Neo4J for GIS and an implementation of GIS by Neo4J

Your question is not very clear.

You say you are trying to create a database of cities and zip codes, but then you say you are using Geonames, which is a database of cities and zip codes. Why are you trying to re-create it?

Places are not hierarchical. You need to think about the institution responsible for the place's boundaries. State boundaries are determined by the US Constitution. City/county boundaries are determined by city and/or state governments. Zip boundaries are determined by USPS.

Some towns contain zips. Some zips contain towns. Some zips cross state boundaries.

A mailing address needs a zip AND some other info. Usually it is a city OR a "surveyed area" such as a street address (which in turn may belong to a city).

A university also likely has more than one address ;)

I would use a little table inheritance and give country, state/prov, city and surveyed area all the same supertype, that way you can point a single foreign key at any of them.