Rethink your query. You have one calculated value from an external source, and one calculated value from an internal source. You are using MAX with no GROUP BY, indicating a single value from each table is to be compared. If the value Datediff(s, '1970-01-01 00:00:00', Max(dob)) that is greater than the value returned by (SELECT Datediff(s, '1970-01-01 00:00:00', Max(dob)) - 3600 FROM [projectuser].[dbo].[sqlquries7]), then all rows from [projectuser].[dbo].[newdataset] will be returned.
That is, if your existing logic worked.
So a definition of what you are actually looking for is the first thing you need to do. This should include what rows you are looking for (i.e. all rows in newDataSet where the dob is after the value calculated from the sqlqueries table.
Note that the way to use the single value from the sqlqueries table is to either (1) place the result in a variable, e.g.:
DECLARE @mdob datetime
SELECT @mdob = Datediff(s, '1970-01-01 00:00:00', Max(dob)) - 3600
FROM [projectuser].[dbo].[sqlquries7]
Or to create a derived table with the single value, and CROSS JOIN it to the other recordset, e.g.:
SELECT *
FROM [projectuser].[dbo].[newdataset] nds
CROSS JOIN (
SELECT Datediff(s, '1970-01-01 00:00:00', Max(dob)) - 3600 AS Cutoff
FROM [projectuser].[dbo].[sqlquries7]) as Tbl
WHERE ( Datediff(s, '1970-01-01 00:00:00', dob) ) >= Cutoff
But remember - MAX in a WHERE clause (or MIN, SUM, etc) is not allowed except in specific, narrowly defined cases. If you need a MAX() value in your WHERE clause, it must be calculated somewhere else.